Mirza Klimenta
Agentic AI for a DeepResearch project
Experience
Agentic AI for a DeepResearch project
Freelance
- Created a multi-agentic system supported by a knowledge graph to automate drafting of research papers
- Used multiple experts (OpenAI models) collaborating during document drafting
- Extracted useful information from the knowledge graph
- Technologies: LangChain, LangGraph, Smolagents, LlamaIndex, dspy
- Infrastructure: Terraform and GitHub Actions (CI/CD) on AWS
- Deployed initial application as a Streamlit app
PulseSpotter
MediaLab Bayern
- Designed to assist journalists in identifying newsworthy topics likely to become popular
- Gathered information from various news sources and analyzed patterns over time
- Suggested emerging trends using AI and machine learning, saving journalists time on research
RAG (LLM)
Freelance
- Led development of an advanced retrieval-augmented generation system to improve HR data retrieval
- Utilized Milvus vector similarity search database and OpenAI API to source and integrate extensive HR-related data
- Attempted to fine-tune an LLM using PEFT (LoRA) and prompt engineering
Senior Data Scientist
Freelancer
Spectral Image Object Detection
Freelance
- Classified signals and identified novelties in spectral images depicting sensor readings
- Relied on self-supervised machine learning paradigms due to insufficient labeled data
- Utilized a YOLO architecture for fast processing using the mmyolo framework
User Grouping in a Social Network
Freelance
- Fostered communication among like-minded users through dynamic group formation based on responses to specific questions
- Employed vectorization and K-nearest neighbors for initial grouping with a recommendation system for refined alignment
Senior Data Scientist
Bayerischer Rundfunk/.pub
- Research on state-of-the-art developments in recommender systems in media (audio, video and textual content)
- Implementation of a recommender system powering ARD Audiothek, resulting in a 15% higher precision than the previous model
- NLP projects: entity recognition and redundancy removal
Fraud Detection for a Buy-Now Pay-Later Platform
Freelance
- Designed and developed a fraud detection model for a Middle-Eastern buy-now pay-later platform
- Employed a graph-based approach during data preprocessing to identify cliques of fraudsters
- Delivered the client’s first successful machine learning project
Next-Best Finance-Related Action
Freelance
- Developed predictive machine learning models to advise individuals on financial actions for long-term goals, such as securing a home purchase
- Context: for an Asian bank offering actionable advice for long-term financial planning
Data Scientist
Yewno/Entropy387
- Research on and implementation of graph-based stock-market prediction models
- Pricing engine implementation achieving a 10% improvement of MAE compared to the previous model
Recommender for Online Betting
Freelance
- Developed a recommender model based on factorization machines to surpass the performance of Amazon Personalize’s hierarchical recurrent neural network
- Improved recommendations through targeted feature engineering
Debt Collection
Freelance
- Predicted likelihood of loan repayment by bank customers to aid segmentation for tailored communication strategies (email, SMS, phone)
- Created a behavioral scoring machine learning model incorporated into Receeve’s collection approach
Sensor Anomaly Detection
Freelance
- Established a baseline machine learning model to predict expected sensor readings and flag anomalies based on deviations
- Enhanced detection accuracy with a graph neural network analyzing connectivity between sensors and observable areas
Post-Doc
Roma Tre University
- Research on graph morphing algorithms
- Implementation of graph drawing algorithms
Pricing Engine for a US Dispatcher
Freelance
- Predicted vehicle transportation prices providing dispatchers with a reliable basis for pricing
- Enhanced model accuracy by approximately 40% through advanced feature engineering and a two-tiered regression strategy combining residual and standard regression with XGBoost
Investment Prediction
Freelance
- Forecasted investors’ decisions on company investments by modeling investors and companies as nodes in a graph
- Developed a graph neural network to predict potential investor-company connections and integrated it into StockFink’s prediction suite
Software Engineer
Visteon (former Johnson Controls)
- (Re)implementation of a testing software
- Implementation of finite state machines
- Code generation
Industries Experience
See where this freelancer has spent most of their professional time. Longer bars indicate deeper hands-on experience, while shorter ones reflect targeted or project-based work.
Experienced in Information Technology (3 years), Media and Entertainment (2.5 years), Banking and Finance (2 years), Automotive (2 years), Education (1 year), and Sport (0.5 years).
Business Areas Experience
The graph below provides a cumulative view of the freelancer's experience across multiple business areas, calculated from completed and active engagements. It highlights the areas where the freelancer has most frequently contributed to planning, execution, and delivery of business outcomes.
Experienced in Information Technology (7 years), Research and Development (6.5 years), Business Intelligence (5.5 years), Product Development (4.5 years), and Finance (2 years).
Summary
Mirza Klimenta received his PhD in Computer Science from the University of Konstanz at age 25. While in academia, Mirza worked in the fields of dimension reduction and graph embedding, and his work has been recognized by the scientific community. As a (Senior) Data Scientist, Mirza focuses on LLMs/RAGs, Recommender Systems, Knowledge Graphs and Classical ML. His most notable work is in the design and implementation of a Recommender System powering ARD Audiothek. He is also a writer of literary fiction.
Skills
Llm
Rag (Graph-rag)
Agentic Ai
Recommender Systems
Deep Learning
Graph Neural Networks
Pricing Engines
Machine Learning
Dimension Reduction
Graph Embedding And Information Visualization
Algorithm Engineering (Large Data Processing)
Numerical Optimization
Programming: Python, Java, C/c++, Matlab
Python: Numpy, Pandas, Scikit-learn, Tensorflow, Keras, Pytorch, Pyg, Networkx, Matplotlib/seaborn, Xgboost/lightgbm/catboost, Spacy/nltk
Recommender Systems: Collaborative Filtering, Two-tower Models, Ensembles, Bandit-based Models, Graph-based Recommenders
Aws: Deployment, Lambda, Ec2, Ecs
Azure: Databricks
Agile Development: Scrum, Kanban
Kubernetes, Streamlit, Spark
Scripting: Shell, Batch
Typesetting: Latex
Tensorflow & Keras: Extensive Experience In Building And Deploying Neural Network Models
Pytorch: Proficient In Model Development And Experimentation, With A Focus On Dynamic Computation Graphs
Pytorch Geometric (Pyg): Skilled In Implementing Graph Neural Networks For Complex Relational Data
Pytorch Lightning: Familiar With This Framework For Scalable And Efficient Deep Learning Model Training
Milvus: Knowledgeable In Building Similarity Search Applications Using This Highly Scalable Platform
Pinecone: Experience In Implementing Vector Search For Machine Learning Models In Production
Athena: Proficient In Serverless Queries With Sql On Large-scale Data Directly In S3
Snowflake: Skilled In Utilizing This Cloud Data Platform For Scalable Analytics
Redis: Experienced In Using This In-memory Database For Caching And Real-time Analytics
Postgresql: Strong Understanding Of Relational Database Management And Development
Aws: Extensive Experience With Sagemaker For Ml Model Development And Deployment; Redshift For Data Warehousing; Lambda For Serverless Computing; Personalize Recipes For Recommender Models
Google Cloud Platform (Gcp): Proficient With Bigquery For Data Warehousing; Developed A Recommender System On This Cloud Provider
Neo4j: In-depth Experience With This Graph Database For Building Knowledge Graphs And Complex Queries
Apache Kafka: Advanced Proficiency In Stream Processing Systems For Building Fault-tolerant, Scalable Real-time Data Pipelines
Apache Druid: Experienced In Real-time Analytics With Druid, Enabling Interactive Queries And Insights On Large-scale Datasets
Grafana: Experienced In Deploying Grafana For Comprehensive Monitoring And Visualization Of Metrics And Logs Across Various Data Sources
Languages
Education
The University of Konstanz
Ph.D., Computer Science · Computer Science · Konstanz, Germany · Magna Cum Laude
The University Sarajevo School of Science and Technology / The University of Buckingham
B.Sc., Computer Science · Computer Science · Sarajevo, Bosnia and Herzegovina · 9.96/10, 98/100
Certifications & licenses
Personalized Recommendations At Scale
Designing State Of The Art Recommender Systems
Profile
Frequently asked questions
Do you have questions? Here you can find further information.
Where is Mirza based?
What languages does Mirza speak?
How many years of experience does Mirza have?
What roles would Mirza be best suited for?
What is Mirza's latest experience?
What companies has Mirza worked for in recent years?
Which industries is Mirza most experienced in?
Which business areas is Mirza most experienced in?
Which industries has Mirza worked in recently?
Which business areas has Mirza worked in recently?
What is Mirza's education?
Does Mirza have any certificates?
What is the availability of Mirza?
What is the rate of Mirza?
How to hire Mirza?
Average rates for similar positions
Rates are based on recent contracts and do not include FRATCH margin.
Similar Freelancers
Discover other experts with similar qualifications and experience
Experts recently working on similar projects
Freelancers with hands-on experience in comparable project as a Agentic AI for a DeepResearch project
Nearby freelancers
Professionals working in or nearby Munich, Germany
